MRpoxNet: A Hybrid Approach for Monkeypox Disease Detection Based on Deep Learning

The monkeypox epidemic has spread to nearly every nation. Governments implement several strict policies, such as forbidding travel, to stop the virus that causes monkeypox. This study used deep learning architectures to identify monkeypox using digitized skin lesion images. Furthermore, a 98.1% accuracy rate was attained using the hybrid MRpoxnet approach. The final five levels of the Resnet 50 architecture were removed in the process of creating the hybrid approach. Five levels were eliminated and ten layers were added in their stead. In the hybrid approach, there are 182 layers instead of the 177 in the Resnet50 design. The accuracy rate has grown even further as a result of these layer modifications made to Resnet50. AlexNet., Resnet50, GoogLeNet., and VGG16 were used for classification and detection. A hybrid approach was also constructed utilizing digital image datasets of monkeypox, other pox disease datasets., and imagess of normal skin. It is therefore concluded from a review of previous scientific works in the literature that the enhanced hybrid approach performs better than alternative deep learning architectures and may be used to computer-aided systems for the diagnosis of monkeypox.